Autonomous Attitude Stabilization of a Quadrotor

Abstract

Quadrotor are four rotor helicopters that fly using a pair of rotors spinning in opposite directions. It is an under actuated system: it uses four different types of controllers: pitch, roll, yaw and altitude to control six degrees of freedom. To obtain the control there is translational rotational coupling. Thus the system is highly non-linear. The aim of the present work is stabilize the quad rotor using a Proportional, Integral and Derivative Controller in real time autonomously. The controller calculates the proximity of the input to the set value, the rate at which the input is moving to the set point and the duration for which the input is away from the set point. These form respectively the proportional, derivative and integral components of the controller and they are weighted and tuned to combine to obtain the motor offset value.